The best open-source image generation, processing, and creative tools available. Every tool below is free, community-maintained, and production-tested. Scoured the earth so you don't have to.
The standard. Text-to-image, image-to-image, inpainting, upscaling. Full web interface with extensions.
Professional creative engine for Stable Diffusion. Industry-leading WebUI, commercial product foundation.
HuggingFace's state-of-the-art diffusion models for image, video, and audio generation in PyTorch.
Run any model (LLMs, vision, voice, image, video) on any hardware. No GPU required. OpenAI-compatible API.
NeurIPS 2024 Best Paper. GPT-style autoregressive image generation. Scaling laws for visual generation.
Stable Diffusion built into Blender. Generate textures, concepts, and reference art inside your 3D workflow.
The universal computer vision library. Image processing, object detection, face recognition, camera calibration.
Fastest Node.js image processing. Resize JPEG, PNG, WebP, AVIF, TIFF at high speed. libvips-powered.
Python Imaging Library fork. The standard for image manipulation in Python. Every format, every operation.
200+ formats. Command-line and API. The Swiss Army knife of image conversion and editing since 1990.
Fast image processing with low memory. Handles huge images efficiently. Powers sharp, imgproxy, and more.
Standalone server for on-the-fly image resizing, processing, and conversion. Fast, secure, production-hardened.
Community-built 2D content creation. Graphic design, digital art, motion graphics. Node-based procedural engine.
Turn doodles into fine art. Seamless texture generation. Semantic style transfer. Example-based upscaling.
Agentic video production. 12 pipelines, 100+ tools, 700+ agent skill files. Full video studio in code.
Remove image backgrounds. One command. Works on anything. Essential for product shots and cutouts.
JavaScript image cropper. Embeddable in any web app. Aspect ratio lock, zoom, rotate, canvas export.
Fast image augmentation. Essential for training data pipelines. 70+ augmentation techniques.
Ready-to-use OCR. 80+ languages, all scripts. Latin, Chinese, Arabic, Cyrillic. Three lines of code.
Add searchable OCR text layer to scanned PDFs. Tesseract-powered. Production-grade output.
Convert images of equations to LaTeX. Vision Transformer model. Math notation to code.
Reusable computer vision tools. Detection, tracking, annotation. Roboflow's gift to the CV community.
Invisible watermark embedding. Extract without original image. Copyright protection for generated art.
Flexible file upload. Drag-and-drop, image preview, crop, resize. Drop-in for any web project.
Content-aware cropping. Finds the best crop rectangle. Used by major CDNs and image processing pipelines.
500+ pretrained segmentation models. UNet, FPN, DeepLabV3. All backbones. Ready to fine-tune.
Convert HTML and CSS to SVG. Generate OpenGraph images from JSX. Vercel's enlightenment library.